(1) Field of the Invention
The present invention relates to a system that inputs and outputs analog electrical signals through computer network links and, more particularly, to an arrangement in which an analog signal input/output device performs analog-to-digital and digital-to-analog conversions.
(2) Description of the Related Art
When voice is input from and output to a personal computer, a driver in the operating system is typically used from an application that handles sound to enable a sound card. In a known arrangement, the sound card is equipped with a speaker terminal and a microphone terminal; the speaker terminal is used to output voice from the computer to the speaker, and the microphone connected to the microphone terminal is used to enter voice.
Particularly, recent personal computers incorporating a CD drive and a DVD drive are widely used to reproduce multi-media information on the computer or to reproduce music contents and video contents downloaded from a network.
Widespread use of the Internet, local area networks (LANs), and the like has also enabled constant connection of a personal computer to the networks and communication with computers and devices on the networks.
When voice is output from the computer in this environment, the above-mentioned arrangement in which devices are connected to the speaker terminal and microphone terminal has restrictions on the installation place and the distance to the computer. Another problem is that a plurality of devices cannot be selectively handled unless a plurality of terminals are provided.
Since home wireless LANs are becoming widespread and freedom in the usage environment, such as for notebook computers, is being increased, devices are preferably connected to the notebook computers via a network.
Printers and scanners, for examples, are already provided with a wireless LAN adapter, so they are widely shared by computers connected to a wireless LAN. All of this type of printers and scanners are devices that handle digital signals having a high affinity with networks and not devices that input and output analog signals.
Voice data may be sent to a remote computer through a network, and the computer outputs the data as voice. This arrangement is used between dedicated terminals in a videophone system or the like. This arrangement is, however, not available for connecting an existing computer to an analog device, so its versatility is low.
General-purpose devices that freely accept analog signals and have them processed by a computer are not provided. There is a desire for a device that can freely input not only voice but also analog signals obtained by measuring instruments and the like into a computer.
In a conventional known example disclosed in Japanese unexamined patent application No. 2003-316375, voice is coded by a terminal and sent to a host computer; the host computer then decodes the coded voice for voice recognition and returns the result to the terminal. Technique to use a general-purpose personal computer as the terminal and have data processing with a high load performed by the host computer, as described above, is known.    Patent Document 1: Japanese unexamined patent application No. 2003-316375
In the above method, however, there is no need to strictly transfer obtained voice data without a delay, so, in practice, any technology for transferring measured data and voice data in real time is not disclosed.
Japanese unexamined patent application No. 2003-163703 discloses a network address translation device that enables heavy voice communications to be established simultaneously between different networks. According to the disclosure, upon receipt of a call setting request, a network address translation (NAT) control unit, which performs address translation, creates a new voice transmission port. The primary purpose of the NAT control unit is to create voice transmission ports on two different networks so that voice communication can be carried out between them, so it is assumed that one socket is used for control and another socket is used for a large amount of voice communication. The two sockets are only selectively used. Various types of control information cannot flow in the control socket during voice communication. When the method of coding analog signals is changed or the method of reproducing the signals is controlled, therefore, a socket dedicated to analog signal transmission must be provided for a signal that changes the coding method or controlling the reproduction method.    Patent Document 2: Japanese unexamined patent application No. 2003-163703
There is also a problem in analog signal input/output; it is difficult to immediately reflect characteristic information indicating the characteristics of the analog signal in the processing at the communication destination. If the analog signal is a voice signal, for example, when a voice level, a sampling rate, the number of bits per sample, and other characteristic information cannot be sent immediately to the device at the communication destination, serious failures may occur during reproduction or recording.
When signals are sent and received via a network, delays may be caused depending on the network situation, so the reliability with which analog signals are processed without a delay is low. Particularly, when stereo voice is simultaneously output from devices on a network, delays may be caused and accumulated separately. When this happens, voice may become completely out of phase.